Third Generation Partnership Project (3GPP) Long Term Evolution (LTE) is the name given to a project in 3GPP to improve the UMTS mobile phone standard to future requirements. Orthogonal Frequency-Division Multiplexing (OFDM) is a digital multi-carrier modulation scheme that is used in LTE. The structure of the OFDM signal in the LTE contains resource elements spaced in time, so-called OFDM symbols, and frequency, so-called OFDM sub-carriers. These resource elements are grouped into a collection of resource blocks that make up the OFDM signal to be transmitted. Within this collection of resource blocks, certain resource elements are designated to contain the control channel signaling information, and base stations within each cell must transmit these control channel resource elements to the various mobiles i.e. mobile terminals, also called user equipment (UE) in LTE, contained within those cells. The transmissions from different cells potentially overlap in either time or frequency and may interfere with each other. Additionally, techniques such as power control may be used for control channel signaling. This affects the level of interference that affects a mobile and it may create an uneven distribution of interference to different mobiles. If certain control channel elements from one base station is transmitted with high power, they might cause disturbance to corresponding control channel elements transmitted from another base station. One technique to overcome this uneven interference scenario is to use interference avoidance, where transmissions are coordinated between base stations, also called NodeB's or eNodeB's in LTE, so that a reduced level of interference is attained at the mobiles.
Alternatively, techniques can be used for making interference appear randomly, which results in that no mobile experiences the same interference pattern repeatedly. In the LTE system, an interference randomization technique is proposed to be used for control channel signaling.
In the suggested approach the control channel elements are interleaved and mapped to LTE transmission resource elements so that there is a randomization of interference between control channels from different cells. A control channel element (CCE) is the control channel information to one or more mobiles. A control channel element group (CCE group) is provided as a concatenation of control channel elements, possibly with a different power level set for each CCE. The CCE group is then mapped to a set of pre-defined control channel transmission resource elements and transmitted. Currently proposed approaches must have a common interleaving scheme for CCE groups transmitted from different cells followed by a cell-specific cyclic shift in order to reduce the number of information symbols transmitted from different CCEs that share common OFDM transmission resource elements, i.e. which create interference between the CCEs. This occurs prior to the mapping to transmission resource elements. The cyclic shift parameter may be tied to the cell identity (ID), for example, so the mobile can easily obtain the cyclic shift parameter. Different interleaving schemes may be used.
Consider the interleaving scheme in an LTE transmission. Let the system have a bandwidth of 5 MHz, so that 24 resource blocks are available for use (note, one resource block consists of 12 OFDM subcarriers spanning over the horizontal dimension and 7 OFDM symbols that span over the vertical dimension as illustrated by the blocks in FIGS. 1, 2, 4 and 5. Let there be two transmit antennas and one to three OFDM symbols used for the CCE group. Interleaving is done in groups of four OFDM tiles i.e. across four adjacent or nearly adjacent resource elements in frequency, so that space-frequency block coding is permitted. Each group of four resource elements is called a symbol group. Symbol groups can alternatively be designed consisting of more or fewer resource elements.
In a configuration where the first three OFDM symbols are used for control channel signaling, there are eight symbol groups per each individual resource block and 192 symbol groups in total across the 24 resource blocks covering the 5 MHz bandwidth.
As an example, FIG. 1 shows the symbol groups (8, 9 and 10) located in one resource block 5 consisting of 12 subcarriers. Only the first three OFDM symbols 7 are shown as these three are potentially used for control channel signaling. Groups of four OFDM tiles make up symbol groups (8, 9 and 10) e.g. the tiles labeled 1 form the first symbol group. The striped and checkered tiles 6 correspond to reference tiles used, for example, for channel estimation and are not available for control or data channel transmission.
A CCE group consisting of 72 symbol groups has been considered for a 5 MHz bandwidth. This might correspond to using all the resource elements in one OFDM symbol, provided that there are no pilot tiles located in that symbol. This corresponds to FIG. 1. The performance of two symbol interleaving patterns has been considered. The patterns are a pruned bit-reversal interleaver. The same interleaver structure is used in all cells and interference randomization is accomplished via a cell-specific cyclic shift of the interleaved pattern prior to the mapping to resource elements.
Within one CCE group, a number of control channel elements are concatenated and transmitted. In the example illustrated in FIG. 1 consisting of 72 symbol groups, there are 9 CCEs each consisting of eight symbol groups.
FIG. 2 shows the concatenation of symbols groups 22 contained within CCE1 20 through the symbol groups 23 contained within CCE9 24 in a control channel element group 21. The eight symbol croups 22 making up CCE1 20 are marked with value 1, while those making up CCE 2 25 and CCE9 24 are marked with value 2 and 9 respectively. In a different cell, if the CCE group to be transmitted has the same format, then two transmissions interfere when the symbol groups from the two CCEs collide with one another. By measuring the number of collisions, it is possible to determine the collision rate performance of the various approaches. Using the interleaving and cyclic shift operations, the amount of interference (i.e. the number of collisions) is potentially reduced. This process is shown in FIG. 3, where the control channel element groups 30 first are grouped together in step 31. The control channel elements are then interleaved in step 32. In step 33 a cell-specific cyclic shift is applied to the interleaved control channel element groups.
The performance for the two interleaving patterns from R1-072225, “CCE to RE mapping”, RAN1#49, Kobe, Japan, May 2007 and R1-072904, “CCE to RE interleaver design criteria”, RAN1#49bis, Orlando, USA, June 2007 is shown in FIG. 10. Performance is evaluated by cyclically shifting the interleaved CCE group and then finding the number of overlapping control channel elements with the same control channel element number, and is the same approach used to evaluate the results in R1-072904, “CCE to RE interleaver design criteria”, RAN1#49bis, Orlando, USA, June 2007. From FIG. 10, the pruned bit-reversal interleaver (PBRI) pattern has high peak correlations when the CCE is shifted by a multiple of nine symbol groups, which the new pattern in R1-072904, “CCE to RE interleaver design criteria”, RAN1#49bis, Orlando, USA, June 2007 avoids these peaks for non-zero shifts.
Consider next a uniformly random interleaving pattern in place of either the two approaches considered in R1-072904, “CCE to RE interleaver design criteria”, RAN1#49bis, Orlando, USA, June 2007. This implies a random permutation of symbol groups before the cyclic shift. To get an idea of the performance under this truly random symbol permutation approach, the mean collision rate for 200 random realizations is shown in FIG. 11. Of course, not all random realization will have adequate frequency diversity, compared to the approaches used in R1-072904, “CCE to RE interleaver design criteria”, RAN1#49bis, Orlando, USA, June 2007. While randomization of the interfering CCE groups reduces interference consistently, a truly random symbol permutation is not practical due to the required signaling aspects between the base station and the mobile for such a scheme.
The difficulty in using the approaches considered in R1-072904, “CCE to RE interleaver design criteria”, RAN1#49bis, Orlando, USA, June 2007 is that they are defined for a specific number of symbol groups mapped to the CCE size and/or the CCE group size. When also accounting for frequency diversity, these approaches are also defined for specific frequency bandwidths. When these parameters change, the interleaving patterns are either no longer valid or may not satisfy the design requirements, i.e. they are not flexible to changing CCE or CCE group sizes, or bandwidth or OFDM symbol allocations in the control channel transmission resources. Consequently, a more flexible approach whose performance approaches the performance of the random realization scheme shown previously is preferred.